Yes — Qwen3.8-27B (27.8B) runs at 55 tok/s on M5 Ultra with 256 GB RAM using Q4_K_M quantization via LM Studio. First token latency is 0.5s. A capable open-source LLM with 27.8B parameters.
Want to run Qwen3.8-27B faster — or step up to bigger models? Find your Mac →The LLMCheck index estimates Qwen3.8-27B on M5 Ultra using our published methodology: Q4_K_M quantization, memory-bandwidth scaling, and cross-referenced third-party benchmarks where available. Figures are transparent estimates — own this config? Submit a real benchmark →
| Metric | Value |
|---|---|
| Tokens per second | 55 tok/s |
| Time to first token | 0.5s |
| Quantization | Q4_K_M |
| Minimum RAM | 256 GB |
| Recommended engine | LM Studio |
| Parameters | 27.8B |
| Benchmark date | 2026-09 |
Q4_K_M 27.8B LM Studio M5 Ultra
The recommended engine for Qwen3.8-27B on M5 Ultra is LM Studio. Install Ollama, then pull the model:
Ollama handles quantization automatically — it will download the Q4_K_M variant (~256 GB) and start an interactive chat session.
| Chip | Speed | First Token | Min RAM | Engine |
|---|---|---|---|---|
| M3 Ultra | 40 tok/s | — | 96 GB | LM Studio |
| M5 Max | 29 tok/s | — | 128 GB | MLX |
| M4 Max | 26 tok/s | — | 64 GB | MLX |
| M3 Max | 19 tok/s | — | 64 GB | MLX |
| M4 Pro | 13 tok/s | — | 48 GB | Ollama |
| M6 | 8 tok/s | — | 24 GB | MLX |
To run Qwen3.8-27B on M5 Ultra you need:
Qwen3.8-27B needs about 256 GB of unified memory. These current Apple Silicon Macs have the headroom to run it comfortably:
As an Amazon Associate, LLMCheck earns from qualifying purchases. These affiliate links cost you nothing extra and help keep the LLMCheck index free.
See how Qwen3.8-27B stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard